Data Analysis Project: E-commerce Reviews Analysis

Welcome to my data analysis project on e-commerce reviews analysis. In this project, I explored customer reviews data from an e-commerce platform to gain insights into customer sentiments and identify key trends. The dataset used for this analysis consists of reviews and associated metadata collected from Kaggle. I named my hypothetical business FZ clothes house.


  • Data collection: The reviews dataset was obtained from Kaggle.
  • Data cleaning: The dataset underwent preprocessing steps to handle missing values, remove duplicates, and standardize text data.
  • Exploratory data analysis: I conducted exploratory data analysis to understand the distribution of ratings, review texts, and other key variables. I also performed sentiment analysis to classify reviews into positive, neutral, or negative sentiments, Product Performance Comparison and Customer Segmentation to analyse the data more.
  • Statistical modeling: I used machine learning techniques, such as text classification algorithms, to predict sentiment based on the review text.
  • Results and interpretation: The analysis yielded insights into customer sentiments, common keywords in positive and negative reviews, and trends in customer feedback over time.

Key Visualizations:

  • Sentiment Analysis:
  • Product Performance Comparison:

  • Customer Segmentation:

Customer Segmentation:

Three customer segments were identified based on their ratings and review text: “Positive,” “Negative,” and “Neutral”

The “Positive” segment consisted of customers who gave high ratings and wrote longer reviews, indicating their strong satisfaction.

The “Negative” segment had negative ratings and average reviews.

The “Neutral ” segment had neither of the positive reviews and negative reviews, suggesting a neutral satisfaction levels.

Insights and Recommendations:

Based on the analysis, here are the insights and recommendations to improve customer satisfaction:

Strengthen Positive Sentiment:

  1. Leverage the positive sentiment expressed by the majority of customers to reinforce their satisfaction.
  2. Highlight positive reviews and testimonials on the company’s website, social media channels, and marketing materials.
  3. Encourage satisfied customers to leave reviews and share their positive experiences.

Address Negative Sentiment:

  1. Pay attention to the small percentage of negative reviews and identify the main pain points or issues raised by customers.
  2. Promptly address and resolve customer complaints or concerns to mitigate any negative impact on overall satisfaction.
  3. Implement measures to proactively gather feedback and address customer issues before they escalate.

Focus on Product Improvement:

  1. Prioritise efforts to enhance the performance of the bottom-performing products (Knits, Blouses, and Dresses).
  2. Analyse customer feedback and ratings specifically related to these products to identify areas for improvement.
  3. Consider product enhancements, quality control measures, or additional customer support to address any identified shortcomings.
  4. Need to change the neutral to positives through implementation of above features

Tailor Communication and Support:

  1. Customise communication and support strategies for different customer segments.
  2. Provide personalised recommendations, offers, and incentives to highly satisfied customers to strengthen loyalty.
  3. Engage with less satisfied customers to understand their concerns and offer solutions or alternatives to address their needs.

Continuous Monitoring and Analysis:

  1. Maintain an ongoing monitoring process to track customer sentiment and satisfaction.
  2. Regularly analyse new customer reviews and ratings to identify emerging trends and address any evolving issues promptly.
  3. Iterate on the analysis periodically to ensure the effectiveness of implemented changes and adapt strategies accordingly.

These recommendations aim to enhance overall customer satisfaction, address specific pain points, and promote continuous improvement based on the insights gained from the analysis.


Through this e-commerce reviews analysis project, we have gained valuable insights into customer sentiments, product performance, customer segmentation, and the acceptance/rejection of the early hypothesis. These insights can help the e-commerce platform make informed decisions to improve product quality, address customer concerns.

Below is a pdf link to the whole process involved.